Abstract: An integrated circuit is presented. The integrated circuit has a set of sensor input interfaces and an output interface. The integrated circuit further has a memory with a first and second memory locations. The integrated circuit further has a multi-chip hub module which has a transaction buffer with both a real-time and a non-real-time buffer. The multi-chip hub module has a context mapper, a re-formatter module and an error handling module. The context mapper is configured to map data to the first or second memory location. The multi-chip hub module is configured to process data through the modules and provide processed data to the output interface.

Abstract: Devices, systems, and methods detect an image frame freeze condition. An example device includes a core logic circuit configured to generate statistics for received image data associated with an image frame, perform a census transform on pixel values of the image data to generate census transformed data, arrange the census transformed data into a binary string having a binary value, and generate transformed image data by replacing a select pixel value of the pixel values of the image data with a decimal value corresponding to the binary value; a load/store engine (LSE) coupled to the core logic circuit, the LSE configured to determine a cyclic redundancy check (CRC) value based on at least one of the image data, the transformed image data, and at least one statistic of the statistics; and an interface configured to transmit the CRC value to a host device.

Abstract: A system is provided. The system generally includes a first processor configured to receive image input data from a red-green-blue infrared (RGBIR) sensor. The first processor of the system is configured to generate a first intermediate image data from the image input data. The system generally includes a second processor. The second processor of the system is configured to generate a second intermediate image data that includes red-green-blue (RGB) image data from the first intermediate image data, and to generate a third intermediate image data that includes infrared (IR) image data from the first intermediate image data. The system generally includes a third processor. The third processor of the system is configured to process the third intermediate image data. The system generally includes a fourth processor. The fourth processor of the system is configured to process the second image data.

Abstract: An image data frame is received from an external source. An error concealment operation is performed on the received image data frame in response to determining that a first frame size of the received image data frame is erroneous. The first frame size of the image data frame is determined to be erroneous based on at least one frame synchronization signal associated with the image data frame. An image processing operation is performed on the received image data frame on which the error concealment operation has been performed, thereby enabling an image processing module to perform the image processing operation without entering into a deadlock state and thereby prevent a host processor from having to execute hardware resets of deadlocked modules.

Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed for image frame freeze detection. An example hardware accelerator includes a core logic circuit to generate second image data based on first image data associated with a first image frame, the second image data corresponding to at least one of processed image data, transformed image data, or one or more image data statistics, a load/store engine (LSE) coupled to the core logic circuit, the LSE to determine a first CRC value based on the second image data obtained from the core logic circuit, and a first interface coupled to a second interface, the second interface coupled to memory, the first interface to transmit the first CRC value obtained from the memory to a host device.

Abstract: A system to implement debugging for a multi-threaded processor is provided. The system includes a hardware thread scheduler configured to schedule processing of data, and a plurality of schedulers, each configured to schedule a given pipeline for processing instructions. The system further includes a debug control configured to control at least one of the plurality of schedulers to halt, step, or resume the given pipeline of the at least one of the plurality of schedulers for the data to enable debugging thereof. The system further includes a plurality of hardware accelerators configured to implement a series of tasks in accordance with a schedule provided by a respective scheduler in accordance with a command from the debug control. Each of the plurality of hardware accelerators is coupled to at least one of the plurality of schedulers to execute the instructions for the given pipeline and to a shared memory.

Abstract: An example device includes: a compute core configured to: send a first request to flash manager circuitry, the first request to store write data in a flash memory; and send a second request to the flash manager circuitry, the second request sent after the first request, the second request to transfer an XIP read operation to the flash memory; the flash manager circuitry configured to: receive the first request; transmit the write data to the flash memory for storing in the flash memory; receive the second request before the storing of the write data is complete; determine whether to preempt the storing of the write data, transmit, in response to a determination to preempt, the XIP read operation to the flash; and the flash memory configured to provide data to the compute core based on the transmitted XIP read operation.

Abstract: Various embodiments disclosed herein relate to hardware enabled pipeline control. In a hardware acceleration system, pipelines are configured to include a hardware enable flag that allows hardware initiation of the pipeline based on triggering of a configurable event. The pipeline can be configured to set the event that triggers the initiation of the pipeline. For example, the end of pipeline of a first pipeline may trigger the initiation of a second pipeline. Accordingly, pipelines that are configured to allow hardware enable based on a specifically configured event are not subject to the extra processing required to initiate the pipeline via software in external memory and triggered by an external controller.

Abstract: In described examples, an integrated circuit includes first, second, third, and fourth image processing blocks, a data selection circuitry, and a pipeline memory. An input of the first image processing block receives raw image data. An input of the second image processing block is coupled to an output of the first image processing block. An input of the third image processing block is coupled to an output of the second image processing block. A first input of the data selection circuitry is coupled to an output of the first image processing block, and a second input of the data selection circuitry is coupled to an output of the second image processing block. A data input of the pipeline memory is coupled to an output of the data selection circuitry, and an output of the pipeline memory is coupled to an input of the fourth image processing block.

Abstract: A data processing device includes a plurality of hardware accelerators, a scheduler circuit, and a blocking circuit. The scheduler circuit is coupled to the plurality of hardware accelerators, and includes a plurality of hardware task schedulers. Each hardware task scheduler is coupled to a corresponding hardware accelerator, and is configured to control execution of the task by the hardware accelerator. The blocking circuit is coupled to the plurality of hardware accelerators and configured to inhibit communication between a first hardware accelerator and a second hardware accelerator of the plurality of hardware task schedulers.

Abstract: Methods and apparatus to extend local buffer of a hardware accelerator are disclosed herein. In some examples, an apparatus, including a local memory, a first hardware accelerator (HWA), a second HWA, the second HWA and the first HWA connected in a flexible data pipeline, and a spare scheduler to manage, in response to the spare scheduler inserted in the flexible data pipeline, data movement between the first HWA and the second HWA through the local memory and a memory. Local buffer extension may be performed by software to control data movement between local memory and other system memory. The other system memory may be on-chip memory and/or external memory. The HWA sub-system includes a set of spare schedulers to manage the data movement. Data aggregation may be performed in the other system memory. Additionally, the other system memory may be utilized for conversion between data line and data block.

Abstract: An image data frame is received from an external source. An error concealment operation is performed on the received image data frame in response to determining that a first frame size of the received image data frame is erroneous. The first frame size of the image data frame is determined to be erroneous based on at least one frame synchronization signal associated with the image data frame. An image processing operation is performed on the received image data frame on which the error concealment operation has been performed, thereby enabling an image processing module to perform the image processing operation without entering into a deadlock state and thereby prevent a host processor from having to execute hardware resets of deadlocked modules.

Abstract: A system to implement debugging for a multi-threaded processor is provided. The system includes a hardware thread scheduler configured to schedule processing of data, and a plurality of schedulers, each configured to schedule a given pipeline for processing instructions. The system further includes a debug control configured to control at least one of the plurality of schedulers to halt, step, or resume the given pipeline of the at least one of the plurality of schedulers for the data to enable debugging thereof. The system further includes a plurality of hardware accelerators configured to implement a series of tasks in accordance with a schedule provided by a respective scheduler in accordance with a command from the debug control. Each of the plurality of hardware accelerators is coupled to at least one of the plurality of schedulers to execute the instructions for the given pipeline and to a shared memory.

Abstract: A method and system for dynamically transferring graphical image processing operations from a graphical processing unit (GPU) to a digital signal processor (DSP). The method includes estimating the number of operations needed for the processing a set of image data; determining the operational limits of a GPU and compare with estimated number of operations and if the operational limits are exceeded; transfer the processing operations to the DSP from the GPU. The transfer can include transferring a portion of executable code for performing the processing operations, and generating a replacement code for the GPU. The DSP can then process a portion of the image data before sending it to the GPU for further processing.

Abstract: Information communication circuitry, including a first integrated circuit for coupling to a second integrated circuit in a package on package configuration. The first integrated circuit comprises processing circuitry for communicating information bits, and the information bits comprise data bits and error correction bits, where the error correction bits are for indicating whether data bits are received correctly. The second integrated circuit comprises a memory for receiving and storing at least some of the information bits. The information communication circuitry also includes interfacing circuitry for selectively communicating, along a number of conductors, between the package on package configuration. In a first instance, the interfacing circuitry selectively communicates only data bits along the number of conductors.

Abstract: An integrated circuit is presented. The integrated circuit has a set of sensor input interfaces and an output interface. The integrated circuit further has a memory with a first and second memory locations. The integrated circuit further has a multi-chip hub module which has a transaction buffer with both a real-time and a non-real-time buffer. The multi-chip hub module has a context mapper, a re-formatter module and an error handling module. The context mapper is configured to map data to the first or second memory location. The multi-chip hub module is configured to process data through the modules and provide processed data to the output interface.

Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed herein to enable data aggregation and pattern adaptation in hardware acceleration subsystems. In some examples, a hardware acceleration subsystem includes a first scheduler, a first hardware accelerator coupled to the first scheduler to process at least a first data element and a second data element, and a first load store engine coupled to the first hardware accelerator, the first load store engine configured to communicate with the first scheduler at a superblock level by sending a done signal to the first scheduler in response to determining that a block count is equal to a first BPR value and aggregate the first data element and the second data element based on the first BPR value to generate a first aggregated data element.

Abstract: In some examples, a system includes storage storing a machine learning model, wherein the machine learning model comprises a plurality of layers comprising multiple weights. The system also includes a processing unit coupled to the storage and operable to group the weights in each layer into a plurality of partitions; determine a number of least significant bits to be used for watermarking in each of the plurality of partitions; insert one or more watermark bits into the determined least significant bits for each of the plurality of partitions; and scramble one or more of the weight bits to produce watermarked and scrambled weights. The system also includes an output device to provide the watermarked and scrambled weights to another device.

Abstract: Methods and apparatus to extend local buffer of a hardware accelerator are disclosed herein. In some examples, an apparatus, including a local memory, a first hardware accelerator (HWA), a second HWA, the second HWA and the first HWA connected in a flexible data pipeline, and a spare scheduler to manage, in response to the spare scheduler inserted in the flexible data pipeline, data movement between the first HWA and the second HWA through the local memory and a memory. Local buffer extension may be performed by software to control data movement between local memory and other system memory. The other system memory may be on-chip memory and/or external memory. The HWA sub-system includes a set of spare schedulers to manage the data movement. Data aggregation may be performed in the other system memory. Additionally, the other system memory may be utilized for conversion between data line and data block.

Abstract: Methods, apparatus, systems and articles of manufacture for an example event processor are disclosed to retrieve an input event and an input event timestamp corresponding to the input event, generate an output event based on the input event and the input event timestamp, in response to determination that an input event threshold is exceeded within a threshold of time, and an anomaly detector to retrieve the output event, determine whether the output event indicates threat to functional safety of a system on a chip, and in response to determining the output event indicates threat to functional safety of the system on a chip, adapt a process for the system on a chip to preserve functional safety.